The unofficial world championships in draughts international 2017

[Krzysztof Grzelak]

Results of the tournament " The Unofficial World Championships In Draughts International 2017 ".

Main round second.

Round 2.9 - 27.04.2017

Flits 3.02 - Kingsrow 1.58 1-1

Dragon 4.5.2 - Moby Dam (Nov 14 2015) 1-1

Edeon Sport 1.1.205 - Scan 2.0 1-1

Aurora Borealis Professional 3.8.2 - engine - Aurora 4 - Sjende Blyn 0-2

Truus voor Windows 1.0 - Tornado 7.07 2-0

Pauses the program Horizon 4.6.

Round 2.10 - 27.04.2017

Tornado 7.07 - Aurora Borealis Professional 3.8.2 - engine - Aurora 4 0-2

Sjende Blyn - Edeon Sport 1.1.205 2-0

Scan 2.0 - Dragon 4.5.2 2-0

Moby Dam (Nov 14 2015) - Flits 3.02 1-1

Kingsrow 1.58 - Horizon 4.6 2-0

Pauses the program Truus voor Windows 1.0.

Round 2.11 - 27.04.2017

Horizon 4.6 - Moby Dam (Nov 14 2015) 1-1

Flits 3.02 - Scan 2.0 1-1

Dragon 4.5.2 - Sjande Blyn 1-1

Edeon Sport 1.1.205 - Tornado 7.07 2-0

Aurora Borealis Professional 3.8.2 - engine - Aurora 4 - Truus voor Windows 1.0 0-2

Pauses the program Kingsrow 1.58.

More information about the tournament - I invite you to the site http://www.chessarbiter.com/turnieje_wa ... 17/f_2697/ The official end of the tournament tomorrow.

[Krzysztof Grzelak]

End of Tournament.

Tournament " The Unofficial World Championships In Draughts International 2017 " ended with the winner of the program Scan of the author Fabien Letouzey - warm congratulations Fabien.

Fabien Letouzey

Scan 2.0

The second place in the tournament took the program Dragon of the author Michel Grimminck - warm congratulations Michel.

Michel Grimminck

Dragon 4.5.2

Third place in the tournament took the program Sjende Blyn of the author Jelle Wiersma - warm congratulations Jelle.

Jelle Wiersma

Sjende Blyn

We also thank the other participants in the tournament.

[center]Official tournament table.[/center]

More information about the tournament - I invite you to the site http://www.chessarbiter.com/turnieje_wa ... index.html

The parties tournament.

[Sidiki]

Very good job.
The level of programs seem to be in progress. The new comer Edeon Sport 1.1.205 play also very Well.
Thank to Krzysztof and the programmers

Sidiki.

[Jelle Wiersma]

My congratulations to Fabien! And my thanks to Krzysztof for running this tournament!
Like Fabien, I'm also surprised by the strength of Edeon Sport; I mean, it's not easy to develop an engine which is stronger than Flits (It took me many years to accomplish that). May be the authors can tell something about it?

[Fabien Letouzey]

My congratulations to Fabien! And my thanks to Krzysztof for running this tournament!
Like Fabien, I'm also surprised by the strength of Edeon Sport; I mean, it's not easy to develop an engine which is stronger than Flits (It took me many years to accomplish that). May be the authors can tell something about it?

I've had a look at its games and it plays a lot like Scan IMO (playing style). That's informal, but in the future I might compute the moves Scan 2.0 would have played in the same positions, for comparison. Also, the analysis scores in the video Sidiki showed are similar to the ones given by Scan 2.0

Prepare for the worst,

Fabien.

[BertTuyt]

I've had a look at its games and it plays a lot like Scan IMO (playing style). That's informal, but in the future I might compute the moves Scan 2.0 would have played in the same positions, for comparison. Also, the analysis scores in the video Sidiki showed are similar to the ones given by Scan 2.0

Prepare for the worst,

Fabien.

Fruit reloaded

l'histoire se répète

Bert

[Fabien Letouzey]

Fruit reloaded

l'histoire se répète

Bert

Bert, I'll be fine. But we will have to take care of this eventually.

For now I prefer to work on the new protocol. I modified both Scan and the Hub GUI, and everything is working. I'm going to write a post about the new syntax; nothing else has changed.

Fabien.

[Krzysztof Grzelak]

For those who are looking for information about Edeon - I invite you to the site http://edeon.ru/article/edeon-100/

[BertTuyt]

For those who are looking for information about Edeon - I invite you to the site http://edeon.ru/article/edeon-100/

I have difficulty understanding the text on the side.
Is there a simple free download for a trial product, or alike.
Or could we have a look at the executable?
If it is based on a Scan copy, I guess it is easy to find out.......

Bert

[Krzysztof Grzelak]

Edeon is a paid program.

[Fabien Letouzey]

I have difficulty understanding the text on the side.
Is there a simple free download for a trial product, or alike.
Or could we have a look at the executable?
If it is based on a Scan copy, I guess it is easy to find out.......

So far I have no reason to think that Edeon is using Scan's search; it would have placed higher in the tournament. Edeon almost lost to Horizon for instance, if I recall correctly. What I am observing is the playing style, which is most influenced (by far) by the evaluation function.

[Sidiki]

I have difficulty understanding the text on the side.
Is there a simple free download for a trial product, or alike.
Or could we have a look at the executable?
If it is based on a Scan copy, I guess it is easy to find out.......

So far I have no reason to think that Edeon is using Scan's search; it would have placed higher in the tournament. Edeon almost lost to Horizon for instance, if I recall correctly. What I am observing is the playing style, which is most influenced (by far) by the evaluation function.

Hi Fabien,

The most propable explaination must be this, some programs like Aurora Borealis have an option that permit to edit manually their opening book. I done a test for you with Scan opening with a lot of his games won in the differents tournaments and dxp games against the other program, Horizon, Kingsrow, Dragon, Dam 2.2, Mobydam, Flits, Truus.
The theory it's that if the sames moves are played, Aurora Borealis will play them perfectly and take advantage. Now if his evaluation it's good it can conlude easily the win if they leave the opening after of course taking advantage.

I think that, i'm of course not a user of Edeon yet,Edeon Evalution function it's good and has perhaps opening of Scan.
This an opening book of Aurora Borealis i edit 2 days ago with Scan moves.
Just remplace it by the existant "international.cbk" on book files
The playing style wil change only when a move will not into the opening that contain almost 500 games.

International.zip

(88.82 KiB) Downloaded 95 times

[Krzysztof Grzelak]

In program Aurora Borealis is no such thing as manual editing of book openings.

[Fabien Letouzey]

The most propable explaination must be this, some programs like Aurora Borealis have an option that permit to edit manually their opening book. I done a test for you with Scan opening with a lot of his games won in the differents tournaments and dxp games against the other program, Horizon, Kingsrow, Dragon, Dam 2.2, Mobydam, Flits, Truus.
The theory it's that if the sames moves are played, Aurora Borealis will play them perfectly and take advantage. Now if his evaluation it's good it can conlude easily the win if they leave the opening after of course taking advantage.

Hi Sidiki,

Let's see if I understand what you mean. You are saying that, maybe, someone built a giant book for Edeon from a large collection of games played by different programs. And, maybe, when I saw the moves I got confused by this?

I don't think so. I analysed Edeon's games from beginning to end. An opening book, no matter how large, cannot cover late moves: there are too many playable moves in draughts. In one of its games against Dragon, Edeon was actually dominating during most of the game. Although it ended in a draw, not many programs can do that ...

You know Scan's style, if you use it a lot. Can you look at Edeon's games and let us know what you think?

Fabien.

[Sidiki]

The most propable explaination must be this, some programs like Aurora Borealis have an option that permit to edit manually their opening book. I done a test for you with Scan opening with a lot of his games won in the differents tournaments and dxp games against the other program, Horizon, Kingsrow, Dragon, Dam 2.2, Mobydam, Flits, Truus.
The theory it's that if the sames moves are played, Aurora Borealis will play them perfectly and take advantage. Now if his evaluation it's good it can conlude easily the win if they leave the opening after of course taking advantage.

Hi Sidiki,

Let's see if I understand what you mean. You are saying that, maybe, someone built a giant book for Edeon from a large collection of games played by different programs. And, maybe, when I saw the moves I got confused by this?

I don't think so. I analysed Edeon's games from beginning to end. An opening book, no matter how large, cannot cover late moves: there are too many playable moves in draughts. In one of its games against Dragon, Edeon was actually dominating during most of the game. Although it ended in a draw, not many programs can do that ...

You know Scan's style, if you use it a lot. Can you look at Edeon's games and let us know what you think?

Fabien.

Hi Fabien

To be honest, the first time i saw the playing style of Edeon i saw immediatly Scan, i know the playing style of all the programs that i use.
I reserved myself to be the first to claim it. The theory that i exposed wasn't necessary the truth, once again, it was a theory.
Now, how is it possible to input Scan's playing style into an engine without have his full strenght. Because like you wrote, it will be placed high at the end of the tournament.
It lose i think 2 or 3 games. This it's strange.